Purpose: Radiographs are commonly used to assess the fit of implant components, but there is no clear agreement on the amount of misfit that can be detected by this method. This study investigated the effect of gap size and the relative angle at which a radiograph was taken on the detection of component misfit. Different types of implant connections (internal or external) and radiographic modalities (film or digital) were assessed. Materials and Methods: Twelve internal-connection and 12 external-connection implant analogs with impression copings were assembled, with radiolucent washers interposed, to produce vertical misfits of 0, 12.7, 25, 38, 51, 63, 76, 88, 102, 114, 127, and 190 µm. A custom-made positioning apparatus was used to obtain radiographs of the components at angulations between 0 and 35 degrees. The images were randomized, and three experienced examiners assessed whether a gap was visible at the interface. Their responses were compared to the actual status of the samples, and a probability model was constructed to predict the likelihood of a correct answer at any combination of gap and angle. Results: The relative angulation of the radiograph and the dimension of the gap were the most significant factors affecting an examiner's diagnostic ability. A 0-µm gap viewed at 0 degrees was the combination most accurately diagnosed. Implant component misfits as small as 12.7 µm were reliably detected with radiographs up to 5 degrees from the orthogonal projection; this was similar with configurations of 25 to 38 µm/10 degrees and 51 µm/15 degrees. There was good (inter-)examiner reliability. Neither the type of component used nor the radiographic media used influenced diagnostic ability. Conclusion: The angulation of the x-ray beam relative to implant components needs to be controlled when using radiographs to detect component misfit. Keywords: dental implants, implant abutment, implant failure, implant misfit, radiographs